Skip to main content

CSV

Recommendation

While CSV is lightweight and easy to use, XML is more well defined and have better support for handling complex data encodings and strctures. We recommend using XML over CSV.

CSV stands for comma separated values. It is a simple format very similar to having data organized in a spreadsheet. The CSV format is not fully standarized even if a proposed RFC4180 exists and different software often interprets the structures differently.

Format​

We expect the following structure (following RFC4180) on CSV files, for a full example please see the CSV tab on our full example

  • First row is a header row

    Example:

    column_header_one,column_header_two,column_header_three
    value_1,value_2,value_3

  • Columns are separated by commas ,

  • Rows are separated by newlines \n

  • Cells that contains newlines or commas must be quoted using double qoutes "

    Example:

    aaa,"my title, which contains a comma",ccc

  • If double-quotes are used to enclose fields, then a double-quote appearing inside a field must be escaped by preceding it with another double quote.

    Example:

    "aaa","b""bb","ccc"

Escaping​

Escaping values in CSV can be tricky, see our section about escaping for more details

Encoding sub-attributes and lists in CSV files​

While the XML format is well suited for sub-attributes (child elements) and lists (repeated elements), the CSV format is not so.

In the CSV format, sub-attributes are added to an attribute as a list inside parentheses. The list of sub-attributes is separated by colon (:). For example, to specify sub-attributes to the shipping attribute:

othercol1,shipping(country:service:price),othercol2
otherval1,SE:standard:10.99 SEK,otherval2

In the CSV format, lists are simply defined as repeated columns with the same name. Here are two shipping services:

othercol1,shipping(country:service:price),shipping(country:service:price),othercol2
otherval1,SE:standard:10.99 SEK,NO:standard:10.99 SEK,otherval2

References​